| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
 
|
|
|
||
|
|
Anesth Analg 2002;95:1012-1018 © 2002 International Anesthesia Research Society
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
Abstract |
|---|
 Top Abstract IntroductionEBM in Clinical PracticeConclusionReferences |
|---|
IMPLICATIONS: This article illustrates the basic principles of evidence-based medicine and the work within the Cochrane Collaboration and the Cochrane Anesthesia Review Group. It describes how important randomized controlled trials and systematic reviews are in providing the best evidence to answer clinically relevant questions.
|
Introduction |
|---|
|
TopAbstractIntroductionEBM in Clinical PracticeConclusionReferences |
|---|
The Cochrane Collaboration is a major focus of activity and a rich source of information within the evidence-based medicine (EBM) movement. The term evidence-based medicine originated at McMaster University in Canada and has been defined as the conscientious, explicit, and judicious use of the best evidence in making decisions about the care of individual patients (1–3). The practice of EBM integrates clinical expertise with the best available external evidence from systematic research.
This article will illustrate the basic principles of evidence-based medicine and the work within the Cochrane Collaboration and the Cochrane Anesthesia Review Group. It describes how important randomized controlled trials and systematic reviews are in providing the best evidence to answer clinically relevant questions.
|
EBM in Clinical Practice |
|---|
|
TopAbstractIntroductionEBM in Clinical PracticeConclusionReferences |
|---|
The practice of EBM has been described by David Sackett (3), and the use of EBM in anesthesiology has been reviewed by Pedersen and Møller (4) and Pronovost et al. (5); however, the following is a short outline of the five principles of EBM.
Ask a Clinical Question You Can Answer
To find an answer to your clinical question, it will need to be clear and well focused. That is: what is the question about? How could you describe a group of patients similar to this particular patient (e.g., an overweight pregnant woman or an elderly person with coronary heart disease)? Which treatment or maneuver are you considering for the patient? (e.g., type of anesthesia, route for central venous catheterization or antibiotic in the intensive care unit). What is the desired (or undesired) outcome for the patient? (reduced postoperative nausea and vomiting, reduced morbidity or mortality, reduced costs).
A good question will cover these three areas of subject, intervention, and outcome, each of equal importance. Failure to frame the question properly will make it hard to find useful and applicable answers. For example, contrast "Is propofol a better induction drug than thiopental?" with "Does propofol lead to fewer unexpected overnight admissions than thiopental after adult day case laparoscopic sterilization?"
Search for Evidence
Information technology has undoubtedly made it easier to search the medical literature, and most institutions provide computers that allow electronic searching. MEDLINE is well known to most doctors, and a subject can be searched by free text and index terms. The Cochrane Library (see full description later in text), EmBase, and Biosis, as well as other databases, can be sought.
Critical Appraisal of the Evidence
Critical appraisal is the process whereby we attempt to answer two main questions about the material we have retrieved in our search. These are 1) are the results important (useful for you as a clinician)? If a study concludes that a diagnostic test is inaccurate or that a treatment has only a negligible effect on patient outcome, there is no need to go through the text to appraise the validity of the study. If however, results are judged to be important, go back and ask 2) are the results valid (close to the truth)? Several questions have to be posed: e.g., was the design of the study sensible? What was the intervention studied and what was it compared with? If the trial was about treatment or prevention, was the assignment of patients randomized, and was the allocation procedure concealed? Were all the patients who entered the trial accounted for at its conclusion, and were they analyzed in the groups to which they were randomized? What outcome was measured and how? Was the outcome assessor blinded? Was a surrogate end point used? Were the statistics used correctly? Finally, were the conclusions that were drawn by the authors supported by the results? Together, these questions help to judge the clinical applicability of the study results.
Clinical Applicability of Evidence-Based Findings
After critical appraisal, some questions remain: how can you integrate this evidence with your own clinical expertise and decide whether and how to incorporate it into the care of your patient? Do these results apply to your patient? Is your patient so different from those in the trial that the results cannot help you? (Can the results be generalized?) Does the intervention match your patients’ (and your) preferences? It is then a clinical decision to estimate the potential benefit of the intervention for the individual patient.
Evaluation
After going through the process of applying EBM in patient care, it is necessary to evaluate the process. How well did we succeed in using EBM? Did we ask the right questions? Did we find useful answers? Will we actually use the findings in clinical practice? Is there anything we can do better?
What Is the Cochrane Collaboration?
The Collaboration is named after the British epidemiologist Archie Cochrane, who wrote: "It is surely a great criticism of our profession that we have not organized a critical summary, by specialty or subspecialty, adapted periodically, of all relevant randomized controlled trials" (6). His criticism is still relevant in that people wanting to make well informed decisions about health care are often confronted with hundreds of thousands of potentially relevant research reports. No one can be expected to sift through these mountains of evidence to discover which forms of health care are more likely to do good than harm.
The Cochrane Collaboration is a worldwide organization. It is composed of 50 collaborative review groups (Table 1), 10 methods groups, 14 Cochrane Centers, and the Cochrane Collaboration steering group. Collaborative review groups are composed of persons from around the world who share an interest in developing and maintaining systematic reviews relevant to a particular health area. Groups are coordinated by an editorial team, who edit and assemble completed reviews into modules for inclusion in the Cochrane Library. The Cochrane Collaboration Steering Group (CCSG) is elected to develop policies and strategies for the Collaboration. It has several subgroups responsible for specific tasks. The CCSG is supported by the Collaboration Secretariat. Methods groups are composed of individuals with an interest and expertise in the science of systematic reviews. They provide advice and support to the Collaboration in the development of the methods of systematic reviews. Each review group is supported by a center. For example, the Anesthesia Group is based in Copenhagen, and its supporting center (the Nordic Cochrane Center) is also based in Copenhagen. The work of Collaborative Review Groups, Methods Groups, Fields/Networks, and the Consumer Network is facilitated in a variety of ways by the work of more than a dozen Cochrane Centers around the world. They share responsibility for helping to coordinate and support members of the Collaboration in areas such as training, and they promote the objectives of the Collaboration at the national level.
|
The first Cochrane Center was opened in October 1992 in Oxford, United Kingdom. One year later, the Cochrane Collaboration was founded as a not-for-profit organization. It was established as a company, limited by guarantee, and registered as a charity in the United Kingdom. The first systematic reviews prepared by the Cochrane Collaboration were by the Pregnancy and Childbirth Group, and these reviews have had considerable influence on obstetric and pediatric practice throughout the world. This has resulted in increased interest from health professionals and government organizations in expanding the scope to other areas of health care. The Cochrane Collaboration’s work is today based on 10 key principles (Table 2).
|
An important contribution of the Cochrane Collaboration is the identification of controlled studies and the creation of a specialized register. This register houses the identified trials that can be accessed to conduct systematic reviews. The reviews prepared within the Collaboration are published in The Cochrane Database of Systematic Reviews and can be revised and updated every 3 mo if necessary. All outcomes from the Cochrane Collaboration are published electronically on CD-ROM and via the Internet. The Cochrane Library gives information on continuing and completed Cochrane reviews. It provides a database of other identified completed reviews, a register of bibliographic information on nearly 250,000 controlled trials, and information about the Collaborative Review Groups (CRGs). The Cochrane Library is widely acknowledged as the best single source of evidence about the effects of health care interventions. It contains The Cochrane Controlled Trials Register, which is now recognized as the most comprehensive bibliography of published reports of controlled trials available.
CRGs
The main work of the Cochrane Collaboration is performed by the CRGs. These are responsible for the preparation and maintenance of Cochrane reviews. A CRG consists of individuals from around the world who share an interest in a particular health topic—in our case, anesthesiology. CRGs are coordinated by an editorial team, comprising a coordinating editor, a number of other editors, and the group coordinator. Other members of the CRG include reviewers (who write and update the systematic reviews), peer referees (individuals not involved with systematic review who offer editorial suggestions before publication), hand-searchers (who search for randomized controlled trials in journals), consumers (people using the prevention or health services), and other interested parties. The primary task of a CRG is to conduct and regularly update systematic reviews of prevention and health care issues within the scope of its group. Each CRG creates a specialized register of methodologically sound controlled studies relevant to their group, of both published and unpublished studies in all languages, avoiding publication bias (which means that the journal appears to favor trials with positive results) (7).
The Cochrane Anesthesia Review Group
The idea of forming the Cochrane Anesthesia Review Group arose in 1997. The Cochrane Anesthesia Review Group (CARG) was established in February 2000. The editorial office is based in Copenhagen, Denmark, and the editorial team consists of people from around the world.1 The main goal of the CARG is to conduct systematic reviews of randomized controlled trials and other controlled clinical trials of interventions. These reviews evaluate outcomes in anesthesia, intensive care, acute pain treatment, prehospital medicine, resuscitation, and emergency medicine.
How to Perform a Systematic Review in Anesthesia?
A review is initially registered with a Cochrane Review Group and Cochrane Title base as a title. That title will become a protocol, which prospectively sets out what is being tested, why, and how it will be done. The complete systematic review adheres to the protocol to maintain uniformity and minimize bias. Systematic reviews performed by the CARG are reviews of studies in which evidence has been systematically searched for, studied, assessed, and summarized according to predetermined criteria.
Titles
To register a title with the CARG, a reviewer submits a registration form (available on the CARG Website, http://www.carg.dk, or on request from the editorial office). The registration form should include contact details for the reviewers, a preliminary title, and a synopsis describing the background, participants, interventions, outcomes, and key words. After the title has been approved by the coordinating editors and the Review Group coordinator has excluded any potential duplication of work or conflicts of interest with other Cochrane groups, the title is registered. The reviewer is then sent guidelines for writing a systematic review ("Tips for Reviewers," available at http://www.cochrane.org/cochrane/hbook.htm), advised to read the Cochrane Handbook and glossary, and sent details of Cochrane training workshops.
Protocols and Reviews
Protocols and reviews are prepared by using the Cochrane Collaboration’s Review Manager software RevMan (which is available on CD-ROM from the editorial office or can be downloaded from http://www.cochrane.org/cochrane/revman.htm). Reviewers who do not have the computer capability to access RevMan should contact the editorial office. The Review Group coordinator acknowledges receipt of the protocol in the editorial office and forwards the protocol, along with guidelines for editing (see http://www.carg.dk: "Tips for Editors/Peer Referees"), to the assigned CARG editor and two peer referees. The editor and peer referees evaluate and comment on the review title, background, objectives, selection criteria, search strategy, methodology, and language of the protocol. The protocol and later the systematic review (the principal output of the Collaboration) will be published electronically in successive issues of The Cochrane Library’s Database of Systematic Reviews. Current reviews and protocols being prepared by members of the CARG are listed in Table 3.
|
Updating
Reviewers are expected to include new trials and update their reviews annually or in response to reader criticism published electronically in the Cochrane Library. The editorial office will provide each reviewer with additional annual references within the scope of the review from the specialized register. The same CARG editor who evaluated the protocol and its resulting review will edit the updated review in the same manner. Reviews that remain unrevised for more than 2 yr will be flagged automatically on the Cochrane Library. If the reviewer does not update the review, it may be reallocated or withdrawn.
What Is a Systematic Review?
A systematic review includes an explicit and detailed description of how it was conducted so that any interested reader should be able to replicate it. The systematic review incorporates strategies to minimize bias and to maximize precision. The systematic review includes a clear research question, criteria for inclusion or exclusion of primary studies, the process used to identify primary studies, the methods used to assess the methodological quality of the selected studies, and the methods used to extract and summarize the results of primary trials on which the conclusions are based (8).
Systematic reviews are evaluated in terms of the following questions: does the review focus on the clinical problem in question? Is the search strategy described in detail, and were all potentially important sources searched? Was the methodological quality assessed and the trials weighted accordingly? How sensitive are the results to the way the review was done? Have the numerical results been interpreted with common sense and due regard to the broader aspect of the problem? The methods for critically appraising systematic reviews in anesthesia have been described recently (4,9).
Categorizing Evidence
Summarized evidence can be categorized according to study design. It should reflect susceptibility to bias. A comprehensive plan for estimating the strength of clinical recommendations can be found on the home page of the Center of Evidence Based Medicine (Oxford) http://cebm.jr2.ox.ac.uk. An adaptation (10) of the North of England Evidence Based Guidelines Development Project (11) (Table 4) describes the categories of quality and strength of evidence in descending order of importance. Clinical questions should be answered by using the best evidence available. If, for example, a question on the effect of an intervention could be answered by Category I evidence, then studies of weaker design (controlled studies without randomization) should not be reviewed. Recommendations are based on consideration of the following: the strength of evidence, the applicability of the evidence to the population of interest, economic considerations, values of the guideline developers and society, and guideline developers’ awareness of practical issues.
|
Why Are Randomized Controlled Trials Preferable in Systematic Reviews?
Randomized controlled trials are quantitative, comparative, controlled experiments in which a group of investigators study two or more interventions in a series of individuals who receive them in random order (12). The real value of randomization is that, if done properly, it reduces the risk of serious imbalance in unknown, but important, factors that could influence the clinical course of the participants. No other study design allows investigators to balance these unknown factors. Proper randomization is the only means by which one can be relatively confident that there are no fundamental differences between groups (concerning known and unknown prognostic factors). When the method of randomization is not described or is open to manipulation (through allocation by birthdate, for example), treatment effect can be overestimated by 30%–50% (13–15). There can be a clear tendency to overestimate effect in the control groups of trials and in studies that are unblinded (13,14). This overestimation can be as much as 30% to 40% (16), or even greater if there are very few patients in the study (17). This discrepancy can be partially offset when observational cohort studies are subjected to strict epidemiological principles. The results can undervalue the treatment effect, which means that the newer statistical techniques do not make the observational studies more reliable (18,19). Some treatments in use today, which are apparently based on good evidence, may actually have either no effect or a negative effect. In the future, more stringent quality controls should be applied to study design, implementation, and reporting of randomization, as well as to the methodology of metaanalyses that include randomized controlled trials.
How can it be determined whether the results of a randomized controlled trial are useful? By looking for information on how the trial was executed, one could judge how well it was executed and whether its results could be used. In a recent publication we stressed that to judge the execution of a trial, one should focus on size, methods, statistics, outcomes, and bias (4).
Other Activities of the CARG
Specialized Register (SR-ANAESTH)
The CARG maintains a register of more than 25,000 randomized controlled trials and clinical controlled trials related to anesthesia, perioperative medicine, intensive care medicine, prehospital medicine, resuscitation, and emergency medicine. The register is maintained with ProCite software, and searches for trials are executed quarterly. Trials included in the register are tagged SR-ANAESTH, and the tag term may be searched in the Cochrane Library. Access to the register is available to reviewers and other members of the CARG.
Hand-Searching
Members of the CARG are involved in searching anesthesia journals by hand to identify randomized controlled trials and clinical controlled trials for inclusion in systematic reviews, the Specialized Register, the Cochrane Library’s database, and the National Library of Medicine. The CARG is currently hand-searching the following journals: Anesthesia & Analgesia, Anesthesiology, Journal of Clinical Anesthesia, Acta Anaesthesiologica Scandinavica, Annales Francaises d’Anesthesie et de Reanimation, Anesthesia and Reanimatology, Canadian Journal of Anaesthesia, Czecko-Slovak Journal of Anesthesiologie, Journal of Cardiothoracic and Vascular Anesthesia, European Journal of Anaesthesiology, Regional Anesthesia, Intensive Care Medicine, American Journal of Respiratory and Critical Care, and British Medical Journal.
Consumer Representation
One of the goals of the Cochrane Collaboration is to make Cochrane evidence accessible to consumers through the Cochrane Library. The CARG is liaising with other review consumer groups to set up good communications and learn how to successfully involve public laypeople who have an interest in or have been subjected to anesthetic care within our group. At present, the CARG has only a few consumers, but we are in the process of collaborating with other consumer groups. More information on this subject is given in the Cochrane Consumers Network’s Website (http://www.cochraneconsumer.com).
|
Conclusion |
|---|
|
TopAbstractIntroductionEBM in Clinical PracticeConclusionReferences |
|---|
The importance of EBM, and the Cochrane Collaboration, has become widely recognized by health professionals and laypeople alike. Progress in applying these principles to anesthesiology, as illustrated by Pronovost et al. (5) and Wijetunge and Baldock (20), has been slow for a variety of reasons. First, hard evidence to support many treatments is simply not available because properly designed studies have not been performed. Second, the evidence may exist but may not be easily accessible to those making the decisions. Third, even when available, the evidence may not be accepted by those delivering care, particularly if it seems to be in conflict with perceived wisdom or personal experience or if it threatens a vested interest.
To overcome these barriers, a collaborative link has been developed between the CARG and Anesthesia & Analgesia. The intention is to disseminate evidence on interventions in anesthesiology currently found in the Cochrane Library to a wider readership. Feedback from readers of Anesthesia & Analgesia will help develop CARG’s future activities.
|
Acknowledgments |
|---|
The Cochrane Anesthesia Review Group and its editorial office are supported by grants from The Copenhagen Hospital Corporation, the National Board of Health, and the Danish Institute for Health Technology Assessment, Copenhagen, Denmark.
|
Footnotes |
|---|
1 The editorial team consists of Jane Ballantyne, United States; Michael Bennett, Australia; John Carlisle, England (Criticisms Editor); Peter Choi, Canada; Jane Cracknell, Denmark (Review Group Coordinator); Martha Delgado, Columbia; Anna Lee, China; Alix Mathieu, United States; Marcus Mullner, Austria; Ann Møller, Denmark (Deputy Coordinating Editor); Nathan Pace, United States (Statistical Editor); Tom Pedersen, Denmark (Coordinating Editor); Andrew Smith, England; Martin Tramér, Switzerland; and Mathew Zacharias, New Zealand.
|
References |
|---|
|
TopAbstractIntroductionEBM in Clinical PracticeConclusionReferences |
|---|
- Sackett DL, Rosenberg WM, Gray JA, et al. Evidence based medicine: what it is and what it isn’t [editorial]. BMJ 1996; 312: 71–2.
[Free Full Text] - Sackett DL, Richardson WS, Rosenberg WM, et al. Evidence-based medicine: how to practice and teach evidence-based medicine. Edinburgh: Churchill Livingstone, 1997.
- Evidence-based medicine: a new approach to teaching the practice of medicine—Evidence-Based Medicine Working Group. JAMA 1992; 268: 2420–5.
[Abstract/Free Full Text] - Pedersen T, Møller AM. How to use evidence-based medicine in anaesthesiology. Acta Anaesthesiol Scand 2001; 45: 267–74.[Web of Science][Medline]
- Pronovost PJ, Berenholtz SM, Dorman T, et al. Evidence-based medicine in anesthesiology. Anesth Analg 2001; 92: 787–94.
[Free Full Text] - Cochrane A. Effectiveness and efficiency: random reflections on health services. London: Nuffield Provincial Hospitals Trust, 1972.
- Crombie IK. Critical appraisal. London: BMJ Publishing Group, 1999.
- Oxman AD, Guyatt GH. Guidelines for reading literature reviews. Can Med Assoc J 1988; 138: 697–703.[Abstract]
- Choi PT, Halpern SH, Malik N, et al. Examining the evidence in anesthesia literature: a critical appraisal of systematic reviews. Anesth Analg 2001; 92: 700–9.
[Abstract/Free Full Text] - Matzen P. Evidensbaseret medicin. In: Lorenzen I, Bendixen G, Hansen NE, eds. Medicinsk kompendium. København: Nyt Nordisk Forlag, 1999: 12–23.
- Eccles M, Freemantle N, Mason J. North of England evidence based guidelines development project: methods of developing guidelines for efficient drug use in primary care. BMJ 1998; 316: 1232–5.
[Free Full Text] - Jaded AR. Randomized controlled trials. London: BMJ Publishing Group, 1998.
- Schultz KF, Chalmers I, Hayes RJ, et al. Empirical evidence of bias: dimensions of methodological quality associated with estimates of treatment effects in controlled trials. JAMA 1995; 273: 408–12.
[Abstract/Free Full Text] - Moher D, Pham B, Jones A, et al. Does quality of reports of randomized trials affect estimates of intervention efficacy reported in meta-analyses? Lancet 1998; 352: 609–13.[Web of Science][Medline]
- Kjærgaard LL, Villumsen J, Gluud C. Reported methodologic quality and discrepencies between large and small randomized trials in meta-analyses. Ann Intern Med 2001; 135: 982–9.
[Abstract/Free Full Text] - Sacks H, Chalmers TC, Smith H. Randomized versus historical controls for clinical trials. Am J Med 1982; 72: 233–40.[Web of Science][Medline]
- Berlin JA, Begg CB, Louis TA. An assessment of publication bias using a sample of published clinical trials. J Am Stat Assoc 1989; 84: 381–92.[Web of Science]
- McKee M, Britton A, Black N, et al. Choose in between randomized and non-randomized studies. In: Black N, Brazier J, Fitzpatrick R, Reeves B, eds. Health services research methods. London: BMJ Books, 1998: 61–72.
- Reeves BC, Machlehose RR, Harvey IM, et al. Comparisons of effects derived from randomized and non-randomized studies. In: Black N, Brazier J, Fitzpatrick , Reeves B, eds. Health services research methods. London: BMJ Books, 1998: 73–85.
- Wijetunge A, Baldock GJ. Evidence-based intensive care medicine [editorial]. Anaesthesia 1998; 53: 419–21.[Web of Science][Medline]
This article has been cited by other articles:
|
|
 |
|
  L.-L. Bo, J.-B. Li, and X.-M. Deng The Assessment of Publication Quality Anesth. Analg., May 1, 2008; 106(5): 1589 - 1590. [Full Text] [PDF]
|
|
|
|
|
|
T. T. Horlocker and D. R. Brown Evidence-Based Medicine: Haute Couture or the Emperor's New Clothes? Anesth. Analg., June 1, 2005; 100(6): 1807 - 1810. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
